Investigating Loss Functions for Extreme Super-Resolution
NTIRE 2020 Perceptual Extreme Super-Resolution Submission.
Our method ranked first and second in PI and LPIPS measures respectively.
Dependency
- Python 3.6
- PyTorch 1.2
- numpy
- pillow
- tqdm
Test
- Clone this repo.
git clone https://github.com/kingsj0405/ciplab-NTIRE-2020
- Download pre-trained model and place it to
./model.pth
.
-
Place low-resolution input images to
./input
. -
Run.
python test.py
If your GPU memory lacks, please try with option -n 3
or a larger number.
- Check your results in
./output
.
Train
- Clone this repo.
git clone https://github.com/kingsj0405/ciplab-NTIRE-2020
-
Prepare training png images into
./train
. -
Prepare validation png images into
./val
. -
Open
train.py
and modify user parameters in L22. -
Run.
python train.py
If your GPU memory lacks, please try with lower batch size or patch size.
BibTeX
@InProceedings{jo2020investigating,
author = {Jo, Younghyun and Yang, Sejong and Joo Kim, Seon},
title = {Investigating Loss Functions for Extreme Super-Resolution},
booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2020}
}